面向航天器装配现场技术问题的知识挖掘和推荐方法

Considering the difficulties in predicting the technical problems and applying the existing knowledge recommendation methods in assembly site due to the characteristics of spacecraft production, the scenario and key factors of technical problems were studied in depth, and a knowledge recommendation method for spacecraft assembly site technical problems was proposed. The organization model of technical problem knowledge instances and problem scenario was proposed to realize the standardized expression of technical knowledge. Then the topic model was used to extract the topic features of the problem scenario, and the key attributes of the technical problems were filtered based on the topic vector of scenario to achieve the quantitative representation of the technical problems. Through the recognition of the characteristics of the current problem scenario, the knowledge recommendation was realized by using the retrieval algorithm. The effectiveness of the method was verified by the experiments on the actual enterprise data set.